GNGTS 2019 - Atti del 38° Convegno Nazionale

GNGTS 2019 S essione 2.2 331 crack model has been extended in order to be able to take into account buckling phenomena of longitudinal reinforcement and effects induced by corrosion. In this work some examples of validation of the PARC _CL model and an application of the model for the definition of Fragility Curves based on the use of non-linear dynamic analyses are presented. The first application is related to the Phase 2 of the CASH benchmark (N005_A469_2014_ EDF_B. TECHNICAL REPORT), an international blind prediction organized by OEDC-NEA (Nuclear Energy Agency). The CASH benchmark was dedicated to the response prediction, under increasing levels of seismic intensity, of multi-storey reinforced concrete (RC) walls used as seismic resisting members in nuclear power plants, Fig. 1. Nonlinear static and dynamic analyses have been carried out to check the reliability of non-linear finite element analysis (NLFEA) to assess the seismic capacity of reinforced concrete walls, characterized by four floors and two bays, to withstand strong earthquakes considered for beyond design situation. The main objective of the CASH benchmark was the evaluation not only of the reliability of finite element analysis but also of the ability, in current engineer practice, to calculate the resistance of RC walls subjected to various seismic loading. The behavior of a “regular” wall and an “irregular” wall 12m width and 16m high have been investigated. The irregular wall differs from the regular one due to an opening located at the third floor of the second bay. Authors attended the benchmark by modelling RC walls using multi-layered shell elements and by adopting PARC_CL 2.0 crack model. Comparison between Phase 2 results obtained from various teams coming from academia, industry and nuclear research organizations showed that the predictions of about an half of the participants were in good agreement. CASH-Phase 2 outcomes prove that today adequate tools for structural analysis are available, though they still have to be improved in order to optimize the nonlinear response prediction. On the other hand, there is a need for NLFEA guidelines (such as those published for static analyses by the Dutch Ministry of Public Works, (Hendriks et al. , 2017; Belletti et al. , 2013b; Belletti et al. , 2014) to reduce the modelling uncertainties. The PARC_CL 2.1 crack model has been also applied in the framework of the SMART2013 international benchmark project in order to obtain fragility curves. The SMART2013 project Fig. 1 - Multi-storey RC wall studied during the CASH benchmark: (a) multi-layered shell elements modelling; (b) non linear dynamic response of the RC wall. Response of the RC base element of the PARC_CL 2.1 crack with: (c) stell reinforcement parallel to the applied load (d) and steel bars oriented at 45°.